TABLE 4.

Select studies assessing the clinical impact of Acinetobacter infection by using comparative control groupsa

ReferenceStudy designCase definition (n)Control definition (n)Method of matching for severity of illness and/or comorbiditiesAttributable mortality (P value) and/or OR (95% CI) in multivariate analysisExcess length of hospital stay (days) (P value and/or adjusted OR [95% CI])Excess length of ICU stay (days) (P value and/or adjusted OR [95% CI])
Abbo et al. (1)Retrospective case control studyIsolation of MDR A. baumannii (infection and colonization) (118)No MDR A. baumannii infection (susceptible infections allowed) (118)McCabe Score15% (P = 0.014 [unadjusted]); adjusted OR, 6.2 (1.3-29.5)6 (P = 0.057)NA
Sunenshine et al. (528)Retrospective matched cohort studyPatients with MDR Acinetobacter infection (96)(i) Patients infected with susceptible Acinetobacter strains (91)Modified Apache III and Charlson score(i) OR, 2.6 (0.3-26.1)(i) 7.7 (P = 0.02; adjusted OR, 2.5 [1.2-5.2])(i) 6.6 (P = 0.04; adjusted OR, 2.1 [1.0-4.3])
(ii) Patients with no Acinetobacter infection (89)(ii) OR, 6.6 (0.4-108.3)(ii) 8.9 (P < 0.01; adjusted OR, 2.5 [1.2-5.4])(ii) 6 (P < 0.01; adjusted OR, 4.2 [1.5-11.6])
Kwon et al. (306)Retrospective matched cohort studyPatients with non-imipenem-susceptible Acinetobacter infection (40)Patients with imipenem-susceptible Acinetobacter infection (40)Pitt bacteremia and Charlson score30% (P < 0.05); adjusted OR, 6.9 (1.8-26.7)NANA
Falagas et al. (159)Retrospective cohort studyPatients given inactive empirical therapy (22)Patients given active empirical therapy (18)Apache II score25.8% (P = 0.10)1.2 (P = 0.53)−3.1 (P = 0.46)
Playford et al. (437)Retrospective case control studyPatients with infection or colonization with carbapenem-resistant A. baumannii (66)Patients without infection or colonization with carbapenem-resistant A. baumannii (i.e., susceptible infections allowed) (131)Apache II score(i) For those with infection, 20% (P = NA) (adjusted OR, 3.9 [1.4-10.7]); (ii) for those with colonization, −18% (P = NA) (unadjusted OR, 0.4 [0.1-1.1])(i) 30 (P = NA) (adjusted HR, 1.7 [1.1-2.7]; (ii) 19 (P = NA) (adjusted HR, 2.2 [1.3-3.5])(i) 15 (P = NA) (adjusted HR, 5.8 [3.3-10.4]); (ii) 6 (P = NA) (adjusted HR, 2.2 [1.3-3.5])
Garcia-Garmendia et al. (181)Prospective, matched case control studyPatients with infection (48) or colonization (27) with A. baumanniiPatients without infection or colonization with A. baumannii (75)Apache II scoreFor those with infection, 43% (P < 0.01); risk rate of death, 4.0 (1.9-8.3)NA13 (P < 0.01)
Blot et al. (40)Retrospective matched cohort studyPatients with A. baumannii bacteremia (45)Patients with no evidence of bacteremia from any cause (90)Apache II score7.8% (P = 0.38); adjusted HR, 0.96 (0.67-1.38)NA5 (P = 0.04)
Garnacho et al. (183)Retrospective matched case control studyPatients with A. baumannii VAP (60)Patients without any type of A. baumannii infection (may have had infection with other organisms) (60)Apache II score11.7% (P = 0.17)NA−1.3 days (P = nonsignificant)
Albrecht et al. (5)Retrospective cohort study involving burn patientsPatients with A. baumannii infection (59)(i) Patients colonized with A. baumannii (52); (ii) patients without infection or colonization with A. baumannii (691)Injury severity scoreFor infected patients vs noninfected patients, 14.3% (unadjusted P < 0.01; adjusted P = 0.65)(i) 12.3 (P < 0.05)b; (ii) 41.8 (P < 0.05)(i) 15.9 (P < 0.05)b; (ii) 26.7 (P < 0.05)
Grupper et al. (215)Retrospective matched cohort studyPatients with Acinetobacter bacteremia (52)Patients without Acinetobacter bacteremia (52)McCabe score36.5% (unadjusted P < 0.01); adjusted HR, 4.01 (1.29-12.53)5 (P = 0.06)NA
Loh et al. (341)Retrospective unmatched case control studyPatients with positive respiratory cultures for A. baumannii (72)Patients with negative respiratory cultures for A. baumannii (91)Not specified4% (unadjusted P = 0.6)10 (P < 0.01)NA
Lortholary et al. (345)Matched cohort studyPatients infected (13) or colonized (27) with A. baumanniiPatients without infection or colonization with A. baumannii (40)Apache II score25%; adjusted RR, 2.0 (1.1-3.6)10.3 (P < 0.01)NA
Wisplinghoff et al. (600)Retrospective matched case control study involving burn patientsPatients with A. baumannii bacteremia (29)Patients without A. baumannii bacteremia (58)No standardized scoring system used to match17% (P = 0.056)NA20 (P = NA)
Choi et al. (88)Retrospective case control studyPatients with Acinetobacter species bacteremia other than A. baumannii bacteremia (28)Patients with A. baumannii bacteremia (112)McCabe score−18.1% (P = 0.13)6 (for A. baumannii controls) (P = 0.02)NA
  • a MDR, multidrug resistant; NA, not assessed; HR, hazard ratio; RR, risk ratio.

  • b Patients infected with A. baumannii had more severe burns and more comorbidities, but excess lengths of stay were not adjusted for these confounding variables.